Represents a thin Python wrapper for implementing Geoprocessing Tools using Rust. A Geoprocessing Tool performs specific tasks on geospatial data. The tools can be integrated in complex geospatial workflows. These workflows are usually designed by Geospatial Analysts in a "Model First" environment.
A custom Geoprocessing Tool can be implemented using Python. But, there is a need for implementing high sophistical geospatial analyses using solid rock runtime environments, libraries and frameworks. As an engineer I do not want to miss the capabilities of
- Efficient code generation (Compiler Buddy)
- Fail fast and often (Compiler says: "No!")
- Pay only for what you use (#YAGNI)
Nowadays, the best option seems to be Rust.
- Create custom Geoprocessing Tools using Rust
pub trait GpTool {
fn label(&self) -> &str;
fn description(&self) -> &str;
fn parameters(&self) -> Vec<GpParameter>;
fn execute(&self, py: Python, parameters: Vec<PyParameterValue>, messages: PyGpMessages) -> PyResult<()>;
}
- Execute any registered Geoprocessing Tool using Rust
let pyresult = gp::tools::execute_tool(py, "arcpy", "ListFeatureClasses", ())?;
let results_as_text = pyresult.as_vecstr();
for result_as_text in results_as_text {
messages.add_message(&result_as_text)?;
}
Building the source by using cargo build release. There is a sample Python toolbox using the provided custom Geoprocessing Tools implemented in Rust.
- Rust v1.44.1
- pyo3 v0.12.3
- ArcGIS Pro v2.6 or later on Windows 10